A. We have conceived and constructed a method for cloning human cDNAs by virtue of phenotypic expression in human cells. This has been done in order to pursue the hormone dependent expression of genes from human breast cancer cells. An initial demonstration that this system works has been carried out in an experiment cloning a fibroblast growth factor gene by examining its expression in an indicator cell line which is absolutely dependent on a fibroblast growth factor activity for anchorage-independent growth. This FGF peptide is a novel peptide produced exclusively by breast cancer cells and is in the process of being sequenced and will be patented. B. In conjunction with studies on fibroblast growth factor peptides in human breast cancer we have examined the expression of the four members of the FGF family, basic FGF, acidic FGF, int-2, and KS- FGF, in human breast cancer cell lines. We have found that the expression of fibroblast growth factor peptides correlates with estrogen unresponsiveness and may be a marker for malignant progression in breast cancer. C. We have completed studies on the action of the ras oncogene in a model system which examines the tumorigenesis of human breast cancer cells in nude mice. We have concluded that activating mutations of the ras oncogene, which markedly diminish its GTPase activity, can enhance the tumorigenesis of this human breast cancer cell line, MCF-7. However, high levels of expression of the ras gene protein, a situation often seen in the clinical breast cancer, does not alter the tumorigenic behavior of these cells.